SECVT Suzuki Motos

Implementation of a Suzuki Electronicallycontrolled Continuously Variable
Transmission in a Formula Student racecar

B.B.M.F. Kuijpers
DCT 2009.015
Bachelor Final Project

Supervisor:
Dr. Ir. Alex Serrarens
Graduate:
Harrie Smetsers
Eindhoven University of Technology
Department Mechanical Engineering
Dynamics and Control Technology Group
Eindhoven, October, 2008
Contents
1 Introduction
2 Suzuki Burgman 650

4
2.1 Driveline Suzuki Burgman 650 ...4
3 Engine selection
5
3.1 Suzuki Burgman 650 engine ....5
3.2 Suzuki GSX-R 600 engine ...6
3.3 Engine choice ...7
4 Suzuki Electronically-controlled Continuesly Variable Transmission (SECVT) 8
4.1 Functioning of a CVT ..8
4.2 Ratio changing mechanism ..9
4.3 Degrees of freedom (DOF) ..10
5 3D CAD model of the SECVT
11
6 Efficiency of the SECVT

12
6.1 Measurement of the efficiency 12
6.2 Race car application .13
7 CVT belt
14
7.1 Bando AVANCE .14
7.2 Limitations of the belt .15
7.3 Contitech Hybrid Ring 16
8 Thrust force of the SECVT
17
9 Cooling of the SECVT
18
10 Control of the CVT

20
10.1 CVT primary pulley position sensor .21
10.2 CVT secondary pulley revolution sensor ..21
11 Measuring of the shift speed
22
11.1 Wiring of the set up ...22
11.2 Calibration of the pulley position sensor (PPS).23
11.3 D-space environment 24
11.4 Problem with secondary pulley revolution sensor 24
11.5 Data processing .24
11.6 Analyzing results ...25
12 Recommendations to achieve higher performance
28
13 Conclusion
30
14 Bibliography
31
A Section of part of Burgman 650 drive line
32
B M-file for calculation of power of modified Suzuki GSX-R 600 engine
33
C.1 Exploded view SECVT
34
C.2 Exploded view primary pulley assembly SECVT
35
C.3 Exploded view secondary pulley assembly SECVT
36
C.4 Documentation of the SECVT
37
C.5 Specifications of the SECVT and Bando AVANCE belt
40
D Efficiency of the SECVT at different ratios
41
E.1 Torque on belt with different reductions
42
E.2 M-file for the calculations of the forces on the CVT belt
44
F.1 Thrust force calculations
48
F.2 M-file for calculation of thrust force
49
G.1 Manufacturing drawing of adapter for test bench
51
G.2 Wiring scheme of the test set up
52
G.3 M-file for calibration of the pulley position sensor
53
G.4 M-file for data processing of the measurement data

while downshifting at 1000 rpm
54
G.5 Test results of measurement shift speed at 2000 and 3000 rpm
59
Chapter 1
Introduction
In 2003 a team was started at the Eindhoven University of Technology to participate in
the international Formula Student competition. The Formula Student competition is an
international design competition in which students from all over the world design and
compete with a single seater race car.
A concept of the transmission might be a continuously variable transmission (CVT). The
advantage of a CVT is that there is no power interruption at the driven wheels unlike with
a conventional stepped gear transmission. This means smooth corner entering and a
higher acceleration than with a conventional gear box, particularly since the engine speed
can be a kept at power-optimal level.
The Suzuki Electronically-controlled Continuously Variable Transmission (SECVT) is a
commercially available CVT which is implemented in a Suzuki Burgman 650 motor
scooter. The CVT is electro-mechanically actuated and uses a v-belt for the transfer of
the engine feasibility for power.
The implementing the SECVT on a race car of University Racing Eindhoven (URE) is
investigated in this Bachelor Final Project. A reversed engineering study should clarify
the functionality of the CVT.
A 3D CAD model of the CVT is produced for a correct documentation of the parts of the
CVT. Besides this, an analysis of the forces on the CVT belt is made to verify if the
strength of the belt is sufficient in the present application. The actuation speed of the
SECVT is tested on an unloaded test bench at the facilities of Drivetrain Innovations
(DTI). Finally some recommendations are made to improve the performance of the CVT.
Figure 1.1: Suzuki Electronically-controlled Continuously Variable Transmission
Chapter 2
Suzuki Burgman 650

The driveline of a Suzuki Burgman 650 is equipped with a Suzuki Electronicallycontrolled Continuously Variable Transmission (SECVT). This CVT uses a set of
variable diameter pulley sheaves. The CVT is electronically controlled with a control
unit. A section of a part of the Burgman 650 driveline is shown in appendix A.
2.1 Driveline Suzuki Burgman 650

The driveline of a Burgman 650 consists of the following components:
Engine
55 hp (40.1 kW) @ 7000 rpm
62 Nm @ 5000 rpm
Primary reduction
Reduction ratio 1.50
CVT
Automatic transmission ratio:
Variable change (1.800 0.465)
Clutch
Wet multi-plate automatic clutch
Centrifugal type
Final reduction
Gear train
Reduction ratio 1.580 (32/31 x 31/32 x 34/31 x 49/34)
Chapter 3
Engine selection
To select the best power source for the race car, different engines had to be compared.
Two different concepts are compared, bearing in mind that the engine for the race car has
some restrictions within the Formula Student competition. The maximum diameter of the
restrictor in the intake is 20 mm and the maximum cylinder volume is 610 cc.
The first concept is the implementation of a Suzuki Burgman 650 engine. The second
concept is a Suzuki GSX-R 600 motorcycle engine. This engine is already used in race
cars of University Racing Eindhoven.
3.1 Suzuki Burgman 650 engine

The specifications of the engine are shown in table 3.1.
Type
Number of cylinders
Power
Torque
Bore
Stroke
Piston displacement
Compression ratio
Fuel system
Starter system
Lubrication system
Four-stroke, Liquid-cooled, DOHC

2
55.00 hp (40.1 kW) @ 7000 rpm
62.00 Nm @ 5000 rpm
75.5 mm (2.972 in)
71.3 mm (2.807 in)
638 cm (38.9 cu. in)
11.2 : 1
Fuel injection system
Electric starter
Wet sump
Table 3.1: Specifications of a Suzuki Burgman 650 engine [ www.bikez.com]
The advantage of this engine is the connection of the engine with the CVT. The
configuration of the Burgman 650 could be adopted.
There are also some disadvantages of this engine, namely:
- The engine produces over a wide rpm range a relative low power. This is not
desirable for a CVT application in a race car.
- Engine not suitable for race applications. The engine is designed for city traffic
and not for a race circuit. The parts of the engine are not designed for high power
production under all circumstances on the race track.
- Cylinder volume of the engine is 638 cc. This exceeds the competition rule of
maximum 610 cc cylinder volume. The engine needs to be modified to conform to
this regulation.
3.2 Suzuki GSX-R 600 engine

The specifications of the original engine and the modified restricted version are shown
below in table 3.2.
Original version
4
115.00 hp (85.8 kW) @ 13000 rpm
69.00 Nm @ 10800 rpm
67.0 mm (2.6 in)
42.5 mm (1.7 in)
599 cm (38.9 cu. in)
12.5 : 1
Electric starter
Wet sump
Type
Number of cylinders
Power
Torque
Bore
Stroke
Piston displacement
Compression ratio
Fuel system
Starter system
Lubrication system
Modified restricted version

4
80.72 hp (59.36 kW) @ 11000 rpm
65.67 Nm @ 8000 rpm
67.0 mm (2.6 in)
42.5 mm (1.7 in)
599 cm (38.9 cu. in)
13.0 : 1
Electric starter
Dry sump
Table 3.2: Specifications of a Suzuki GSX-R 600 engine [ www.bikez.com]
The power of the modified engine is measured at the engine test bed of University Racing
Eindhoven. An M-file is created (appendix B) to calculate the power of the engine at the
crankshaft. An assumption of a drive train efficiency of 92 % is taken into account,
because the power is measured at the driveshaft. Figure 3.1 shows the power and torque
of the engine.
Power of modified Suzuki GSX-R 600 engine at crank
60
40
40
Torque [Nm]
Power [kW]
60
20
20
0
3000
4000
5000
6000
7000
8000
9000
10000
11000
12000
13000
RPM
Figure 3.1: Figure of modified Suzuki GSX-R 600
The advantages of this engine are:

- High power of the engine, even with the restriction.
- Lot of knowledge of the engine in the team.
- Dry sump lubrication system (modified version). With the high lateral forces on
the car, a dry sump system is more suitable for a race car application. This ensures
there is oil pressure under all circumstances. With the system, the height of the
engine is reduced which is beneficial for the center of gravity of the car.
- Suitable for race applications. The engine is used in several race competitions
over the world.
A disadvantage of this engine choice is the unfamiliar connection of the CVT to the
engine. This needs be designed and manufactured. Another problem could be the higher
power (60 kW instead of 40 kW) that needs to be transmitted with the CVT belt. This
could mean failure or excessive wear of the belt.
3.3 Engine choice

The Suzuki GSX-R 600 engine is chosen over the Burgman 650 engine for the driveline
of the race car. The high power output and the experience with the engine are the reasons
for this choice.
The necessary modifications of the Burgman 650 engine to conform the regulation for the
competition where also a problem in the short time period of the design of the car.
Chapter 4
Suzuki Electronically-controlled Continuously

Variable Transmission (SECVT)
4.1 Functioning of a CVT
The principle of a CVT is transmitting torque by friction (friction between the pulley
sheaves and the belt) over an infinite number of ratios. By squeezing the belt between the
pulleys, torque can be transmitted by tension or compression forces in the belt. The
SECVT is equipped with a tension belt.
The ratio of a belt CVT is defined by the pitch radius on which the belt runs on each of
the pulleys. The position of the pulleys can be adjusted gradually between the minimum
and maximum setting, every possible ratio between the minimum and maximum setting
can be achieved and used.
Secondary
Fixed Pulley
Primary Slide
Pulley
Primary Fixed
Pulley
Secondary
Slide Pulley
Input Shaft
Output Shaft
Figure 4.1: Top view of SECVT
Figure 4.2: Primary pulley assembly (open)
Figure 4.3: Secondary pulley assembly (closed)
4.2 Ratio changing mechanism

The ratio of a CVT is controlled by changing the distance of the pulley sheaves and thus
the radius over which the belt runs. Moving a pulley sheave inwards or outwards is
achieved by actuating the movable pulley sheaves.
The SECVT uses an electric motor that actuates the primary slide pulley sheave in an
electro-mechanical way. The secondary pulley will follow the displacement of the
primary pulley because of a spring that is incorporated in the design of the secondary
pulley. The primary pulley is fitted with a spindle which is driven by a DC-motor and
two plastic gears (figure 4.4). The total gear reduction for this mechanism is 1: 90.85; the
spindle involves a reduction by altering a rotational movement into a translating
movement. The size of this reduction causes the ratio changing system to be self
breaking, meaning if the DC-motor is not powered, the spindle will maintain its position
and thus the ratio of the SECVT remains the same.
DC-motor
Gear
reduction
Spindle mechanism
(actuation)
Translation of
pulley sheave
Figure 4.4: Actuation mechanism SECVT
As mentioned before, the secondary pulley (figure 4.5) is equipped with a spring which
allows this pulley to follow the ratio change of the primary pulley. The spring in the
secondary pulley also ensures the clamping force from the pulley sheaves to the belt
(ratio dependent). The torque-cam mechanism of the secondary pulley is able to increase
the clamping force if a torque peak is detected from the output shaft of the SECVT. The
torque cam inside this pulley allows the pulley to increase the clamping force is enlarged
due to the slope and friction material on the sliding surface. This mechanism allows the
SECVT to be controlled fairly simple.
Compression
spring
Torque-cam
mechanism
Figure 4.5: Torque-cam mechanism secondary pulley SECVT
4.3 Degrees of freedom (DOF)

Every part of the SECVT has its own degrees of freedom. Below is explained what the
DOF are for the different components (component shown in red) (DOF in green).
Primary pulley
- Primary fixed pulley
o Rotation around x-axis
Primary slide pulley

o Translation along x-axis
o Pulley sheaves rotate at same speed
y
z
-
Primary slide pulley gear (when actuated)

o Translation along x-axis
Screw shaft
o Fixed to housing
y
z
Secondary pulley
- Secondary fixed pulley
y
-
Secondary slide pulley

o Translation along x-axes
o Pulley sheave can rotate independent
within the limitations of torque-cam
mechanism (chapter 8)
y
10
Chapter 5
3D CAD model of the SECVT

To study to functioning of the CVT, it is disassembled at the workshop of University
Racing Eindhoven. A 3D model of the SECVT is made with the computer program UGS
NX 5.0. The specifications of the specific parts are documented in appendix F. The
number of the parts in appendix C.4 is related to the numbers in appendix C.1, C.2 and
C.3.
Figure 5.1: Side views of 3D CAD model of SECVT
11
Chapter 6
Efficiency of the SECVT

In order to get a high performance driveline of the car it is important to know the
efficiency of the CVT. At the Technical University of Eindhoven (TU/e) experimental
research is performed to determine the efficiency of the SECVT [Study on the efficiency
of an A-CVT].
6.1 Measurement of the efficiency

The CVT is tested for certain ratios on a loaded test bench. Torque and angular velocity
is measured at the in- and output shaft of the CVT. After data processing the efficiency of
the transmission can be calculated by using equation 6.1.
P
transmission = out
Equation 6.1
Pin
Pin = primair T primair
Equation 6.2
Pout = sec ondair Tsec ondair
Equation 6.3
This will give the efficiency of the system. The results of these measurements, more
specific the efficiency of the transmission can be seen in figure 6.1.
Figure 6.1: Efficiency of SECVT at ratio low
12
In figure 6.1 the efficiency is shown for the lowest CVT ratio. In this case the efficiency
is between 92 % and 95% when the primary torque is more than 20 Nm.
In the application of a race car the primary torque is always higher than 20 Nm when
power is applied on the CVT.
For other ratios the efficiency is shown in appendix D. It shows that the efficiency of the
CVT is about 90% for most cases. For medium ratio the efficiency reaches even 95% for
the majority of the torque span. At overdrive ratio the efficiency drops, because of higher
friction in bearings and over clamping of the belt (chapter 8).
6.2 Race car application

On track test results have to show whether the efficiency of the SECVT is high enough to
achieve higher performance than the stepped gear transmission in the race car.
13
Chapter 7
CVT Belt
The belt which is used in the SECVT is a design of Bando Chemicals and is developed as
the next generation v-belts for heavy duty use in automotive applications and is called the
Bando AVANCE.
Figure 7.1: Section of the Bando AVANCE belt
7.1 Bando AVANCE

The belt consists of blocks and tension members which are displayed in figure 7.1. Since
the dry hybrid v-belt is used as a pull belt, the transmission of power from the primary
pulley to the secondary pulley is done by a surplus of pull forces in the belt. When
pulling the blocks alone, no power can be transmitted. This means that the force has to be
directed through the blocks to the belt at the primary pulley. This is done by creating
notches in the tension members in which the blocks grip the tension member and transmit
the force and vice versa at the secondary pulley.
The blocks are made of an aluminum core since they have to transmit power from the
pulley sheave to the belt. This means that a certain thrust force has to be used to do so.
The aluminum core of the blocks ensures the strength of the belt when clamping the belt
while the aramid cords ensure the tensile stiffness of the belt. The required clamping
force on the belt is limited because a friction material is added to the blocks and to the
surface of the belt. This and the fact that no lubrication oil is used, causes the friction
coefficient to increase and the required thrust to decrease at a certain transmittable torque.
Therefore the required thrust on the dry hybrid belt is not as high as the thrust force on a
conventional metal belt that is combined with an oil lubricated environment.
14
To improve the contact between pulley surfaces and the belt even more, the pulleys are
treated with non-electrolyte Ni-P plating so that a good friction coefficient, abrasion,
resistance and rust resistance is achieved. This layer causes the pulley and the friction
coefficient to be sensitive to damage or wear. When damage is inflicted to this surface, it
can be assumed that the torque cam mechanism is no longer capable of applying
sufficient thrust force to the pulley-belt contact to transmit the applied power to or from
the belt.
7.2 Limitations of the belt

To ensure a good lifetime of the belt, the manufacturer has introduced some limitations of
the belt. The maximum speed of the belt is approximately 40 m/s and the maximum
applied torque is 135 Nm (appendix C.5).
When the SECVT is used in combinations with the Burgman 650 engine, these
limitations are not a problem. In the case that the modified Suzuki GSX-R 600 engine is
applied the limitations could be a problem. When the speed of the belt is more than 40
m/s there can be assumed that the belt will slip (> 1%) between the pulley surfaces. This
increases heat development in the belt which cause more wear of the belt.
To ensure that the speed of the belt will be below the limitations, the optimal primary
reduction is 2.2724 (out/in). In this case the maximum speed of the belt is 40 m/s and the
maximum applied torque on the belt is above 140 Nm. This will probably cause more
wear to the belt, but not a failure of the belt. Manufacturers normally apply safety factors
to ensure a long lifetime of the product. Therefore we trade higher torque transfer with
lower lifespan of the belt. It has to be researched what the lifetime reduction will be in
case of (sustained) overloading. At maximum input power, however, the torque is
substantially lower than 135 Nm.
Speed of the CVT belt @ 13000 RPM
60
max allowable speed
with primary reduction
with primary reduction and 5th gear
with primary reduction and 6th gear
with ideal reduction
55
50
speed belt [m/s]
45
40
35
30
25
20
15
10
0.035
0.04
0.045
0.05
0.055
0.06
0.065
radius primary pulley [m]
Figure 7.2: Speed of the CVT belt at different primary reductions
15
Torque on CVT belt with ideal reduction

200
180
160
Torque on CVT belt [Nm]
140
120
100
80
60
40
Torque on belt
Max torque Bando Avance
Max torque Contitech Hybrid Ring
20
1500
2000
2500
3000
3500
4000
4500
5000
5500
rpm primary pulley
Figure 7.3: Torque on the CVT with the ideal primary reduction
7.3 Contitech Hybrid Ring

Contitech (division of Continental) has announced a new type of hybrid belt. This belt
has the same dimensions as the Bando AVANCE, but the maximum allowed torque is
much higher. The permitted torque is 100 to 130 Nm. The belt can withstand torque
peakes up to 180 Nm. This while the efficiency of the CVT (belt and pulley sheaves) is
90-95 %. The belt is not yet commercialy available (only for OEM evaluation).
Figure 7.4: Contitech Hybrid Ring

[www.contitech.de]
Figure 7.5: Section of Contitech Hybrid Ring

[www.contitech.de]
16
Chapter 8
Thrust force of the SECVT

The thrust force in the SECVT is generated by two components namely the spring and
the torque-cam in the secondary pulley. The primary pulley forces the belt to run on a
certain running radius at the primary pulley as on the secondary pulley. This implies that
each speed ratio cause a fixed thrust force of the spring. This thrust force is increased by
making use of the torque-cam. At a certain load the cam will press against the follower,
causing an additional thrust force on the belt.
Figure 8.1: Forces of torque cam in secondary pulley
The graph in figure 8.1 shows the thrust force in relation to the applied torque and ratio.
The calculations of the thrust force can be found in appendix F. The additional thrust
force generated by the torque-cam is only dependant on the applied torque and not of the
present ratio. This could mean that the safety factor in high gear will be very high when
transmitting a low torque. This forces the efficiency to decrease when transmitting a fixed
input torque and speed while shifting towards high gear.
The modified Suzuki GSX-R 600 engine produces a lot more torque in comparison to the
original Burgman 650 engine. This causes that the thrust force is a lot higher in
comparison to the Burgman 650 engine. When over clamping occurs energy is dissipated
in the belt and thus not reaching the output shaft of the CVT. This could mean a lower
efficiency of the CVT or increased wear of the belt. Extensive testing of the CVT on a
loaded test bench could evaluate this.
17
Thrustforce secondary pulley SE-CVT with torque cam

7000
Torque at input shaft
0 Nm
6000
20 Nm
40 Nm
Modified Suzuki GSX-R 600
74 Nm
149 Nm
5000
Thrustforce [N]
Burgman 650
4000
3000
2000
Spring (0 Nm)
1000
0.6
LOW
0.8
1.2
1.4
1.6
1.8
CVT ratio [-]
OD
Figure 8.2: Thrust force of the secondary pulley on CVT belt
18
Chapter 9
Cooling of the SECVT

The efficiency of the SECVT is less than 100%. The energy that is not transmitted with
the CVT belt will be converted into heat. To ensure that the heat will be diverted swiftly
to the surroundings, a cooling system is implemented.
A build-in fan will create airflow in the SECVT. An air duct at the front side of the
Burgman 650 motor scooter channels air to the fan of the CVT. Figure 9.1 shows
schematically the flow that is created with the fan. A cochlea circulates the air flow in the
CVT casing and pushes the heated air out of the CVT casing.
Figure 9.1: Schematically representation of the flow in the CVT casing
19
Chapter 10
Control of the CVT

The SECVT is electronically controlled with a control unit. This control unit is
incorporated in the ECU of the Burgman 650. The unit controls the CVT motor which
actuates the primary slide pulley.
Different sensors are connected with the control unit. Figure 10.1 is a schematic
presentation of the functioning of the CVT control unit. The CVT primary pulley position
and CVT secondary pulley revolution sensor are incorporated in the casing of the
SECVT.
Figure 10.1: Overview of the functioning of the control unit of the SECVT
20
10.1 CVT primary pulley position sensor

For the control strategy of the ratio change mechanism, a pulley position sensor is used to
measure the presence of slip in the CVT. This is necessary since the dry hybrid v-belt is
not allowed to slip for more than 1% over a period, longer than a few seconds. If this
happens, permanent damage can be done to the belt. The pulley position sensor is nonlinear axial displacement sensor. The contact surface between the sensor and the pulley
does rotate but only when the ratio is changed and even then at a low rotational speed.
This means that the influence of this rotating surface is a low and that the contact is
maintained which allows a correct read out of the sensor. The contact between the sensor
tip and the gear is ensured by a spring which is incorporated in the design of the sensor.
Specifications sensor:
- non-linear axial displacement sensor
- 0.06 V sensor voltage 5.04 V
- Input voltage: 4.5-5.5 V
- Resistance: Compressed: 1.9-2.3 k
Extended: 0.2-1.0 k
Figure 10.2: Primary pulley position sensor
10.2 CVT secondary pulley revolution sensor

The speed sensor which is used in the transmission for control purposes is an inductive
pulse sensor which is combined with 18 notches in the secondary pulley. A speed sensor
at the primary pulley is not available in the transmission since the primary pulley speed is
obtained by measuring the crankshaft speed and multiplying it with gear reduction in
between.
Specifications sensor:
- inductive pulse sensor
- 18 notches on secondary pulley
- input voltage: > 7 V
- resistance: 400-600
- sensor peak voltage : > 5 V (idle speed)
- effective range: 0.38 3.62 V
Figure 10.3: Secondary pulley revolution sensor
21
Chapter 11
Measuring of the shift speed

To determine if the shift speed of the SECVT is sufficient for the application of a race
car, measurements are done. The shift speed of the CVT is measured on an unloaded test
bench at Drivetrain Innovations (DTI) (figure 11.1).
The test bench consists of an electric motor which is capable of maximum 3000 rpm. The
maximum allowed revolutions per minute of the CVT is higher as this value, but the
specification of the electric motor is sufficient to test the functioning of the CVT.
The outgoing shaft of the electric motor is connected to the CVT with an adapter
(appendix G.1) which is custom made at VDL.
The data acquisition that is used for the measurements is D-space.
Electric motor
SECVT
Adapter
Figure 11.1: SECVT on unloaded test bench of Drivetrain Innovations
11.1 Wiring of the set up

The first step of building the measuring set up is to connect the computer interface with
the sensors of the test bench. A schedule of the wiring of the set up can be seen in
appendix G.2.
22
11.2 Calibration of the pulley position sensor (PPS)

To achieve correct test results, the PPS has to be calibrated. The displacement of the
sensor is measured with a micrometer. At the same time the outgoing voltage of the
sensor is measured. The measurements are done three times, to ensure that the calibration
is correct (table 11.1).
With Matlab a polyfit is made (appendix G.3). The average data is used in a look up table
in the D-space environment.
Compression Measurement 1 [V] Measurement 2 [V] Measurement 3 [V] Average [V]

PPS [mm]
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
0,142
0,142
0,235
0,226
0,358
0,347
0,485
0,478
0,632
0,623
0,771
0,747
0,911
0,883
1,06
1,03
1,21
1,18
1,35
1,34
1,55
1,53
1,75
1,73
1,97
1,96
2,2
2,18
2,43
2,41
2,73
2,72
3,03
3,05
3,39
3,4
3,81
3,8
4,32
4,3
4,82
4,82
Table 11.1: Calibration of the PPS sensor
0,142
0,241
0,364
0,491
0,628
0,777
0,991
1,06
1,19
1,34
1,53
1,76
1,98
2,19
2,46
2,72
3,04
3,4
3,8
4,28
4,82
0,142
0,234
0,356
0,485
0,628
0,765
0,928
1,050
1,193
1,343
1,537
1,747
1,970
2,190
2,433
2,723
3,040
3,397
3,803
4,300
4,820
23
11.3 D-space environment

In D-space an environment is created to control the DC motor of the SECVT and to
obtain the sensor outputs of the test rig. The environment is shown in figure 11.2.
10
position_v oltage
position_mm
Look-Up
T able
Bad Link
Scope2
DS1102ADC
Bad Link
interrupt
Hardware Interrupt
Trigger()
In1
Out1
aantal_nokken
notch_count
Pul seCount
0
Engine_Start
1/10
1/18
-K -
T ransport
Delay
60
1
0.01s+1
T ransfer Fcn
rpm
Scope1
Scakelm otor
Bad Link
DS1102DAC
Figure 11.2: D-space environment
11.4 Problem with secondary pulley revolution sensor

During the test a problem encountered with the secondary pulley revolution sensor. Dspace could not process the data of the sensor. Warnings were shown on the screen,
because there were problems with overflows. Measurement showed that correct data was
received in D-space, but when the data was logged to an outgoing file, problems
occurred. There is decided to ignore the problem and finish the test without the data of
the sensor.
11.5 Data processing

An M-file is created to process the received data (appendix G.4). Because of the problem
with the secondary pulley revolution sensor, the rpm of the secondary pulley is calculated
with the data of the primary pulley position sensor. With the data of the primary pulley
position sensor the CVT ratio is calculated. Together with the rpm of the primary pulley,
the rpm of the secondary can be determined.
24
11.6 Analyzing results

The final step is analyzing the test results. The revolutions per minute of the primary
pulley are remained constant at 1000, 2000 and 3000 rpm. Every measurement is done
five times, to verify if the results change per measurement. Figure 11.3 shows the
compression and voltage of the pulley position sensor. Figure 11.4 shows the rpm of the
primary and secondary pulley while the CVT shifts down in its ratio. You can see that the
rpm of the primary pulley is constant (1000 rpm) and the rpm of the secondary pulley is
changing. The time that the CVT shifts down in its ratio change is approximately 1.45 s.
This value does not change for different rpm of the primary pulley (table 11.2). The
figures of the data of the measurements at 2000 and 3000 rpm of the primary pulley can
be found in appendix G.5. The figures in the appendix show that the shift time does
change while shifting up for different motor speeds of the primary pulley (see also table
11.2).
An explanation is that a higher force is required to move the secondary pulley against the
direction of the compression spring. While downshifting this works in the opposite
direction and it is an advantage. Besides that a race car decelerates faster than it
accelerates, so a shorter downshifting time is good for the implementation of the SECVT
in a Formula Student race car.
Another thing that changed was the current of the power supply of the DC motor (peak
values of 20 A). When the pulleys rotate at higher speeds it is easier to alternate the ratio.
Therefore the current of the CVT motor drops at higher rotational speeds while the
voltage is constant. What can be expected out of the measurements is that the current of
the CVT motor will drop more if the input speed is higher. The nominal input speed in a
Formula Student race car is around 5500 rpm. This also decreases the shift time of the
CVT.
Compression and voltage of the pulley position sensor while downshifting at 1000 rpm
22
compression
voltage
20
18
compression [mm] / voltage [V]
16
14
12
10
0.5
1.5
2.5
3.5
time [s]
Figure 11.3: Compression and voltage of the PPS while the CVT shifts down at 1000 rpm
25
Revolutions per minute of the primary and secondary pulley while downshifting at 1000 rpm
2500
primary pulley
secondary pulley
2000
rpm [-]
1500
1000
500
0
0
0.5
1.5
2.5
3.5
time [s]
Figure 11.4: RPM of the primary and secondary pulley while the CVT shifts down at
Compression and voltage of the pulley position sensor while upshifting at 1000 rpm
22
compression
voltage
20
18
16
14
12
10
0.5
1.5
2.5
3.5
time [s]
Figure 11.5: Compression and voltage of the PPS while the CVT shifts up at 1000 rpm
Note: In the final phase of down shifting of the CVT, the belt looses the
contact with the pulley sheaves. The result of this is shown in figure 11.5
and 11.6. The lines of the measurements are not exactly the same.
26
Revolutions per minute of the primary and secondary pulley while upshifting at 1000 rpm
2500
primary pulley
secondary pulley
2000
rpm [-]
1500
1000
500
0.5
1.5
2.5
3.5
time [s]
Figure 11.6: RPM of the primary and secondary pulley while the CVT shifts up at 1000 rpm
RPM primary pulley

1000 (down)
1000 (up)
2000 (down)
2000 (up)
3000 (down)
3000 (up)
Shift time [s]

1.47
2.14
1.45
1.93
1.45
1.86
Table 11.2: Shift times of the SECVT at different motor speeds

Trend of shifting speed
3
Downshifting
Polyfit downshifting
Upshifting
Polyfit upshifting
2.5
time [s]
1.5
0.5
0
500
1000
1500
2000
2500
3000
3500
rpm primary pullley [-]
Figure 11.6: Shift speeds plotted for different rpm of the primary pulley
27
Chapter 12
Recommendations to achieve higher

performance
In order to achieve even a higher performance of the CVT, some modifications could be
made. This can improve the reliability and the performance of the race car. The following
recommendations are suggestions. None of them are tested in reality.
CVT casing
The weight of the casing can be reduced. This can be achieved with modifying the
original casing with milling or a design of a new casing of other material. A material such
as carbon fiber can save approximately two third of the weight. With this material is a
complex production method is required.
Since the CVT variator has a dry belt and mechatronic actuation, the casing does not need
to be entirely closed, which means that holes can be drilled in non critical areas.
Pulley sheaves weight reduction
In order to reduce the overall weight and inertia of the pulley sheaves, the pulley sheaves
can be modified. Part of the pulley sheaves can be removed (figure 12.1). Another option
is to redesign the pulley sheaves of another material. With the use of lightweight
aluminum a lot of weight can be saved. The aluminum can be treated with non-electrolyte
Ni-P plating so that a good friction coefficient, abrasion and resistance can be achieved
[www.coatingservice.nl]. A FEM analysis could prove the strength of the pulley sheaves,
so no bending occurs.
Figure 12.1: Concept for reducing the weight of the pulley sheaves
28
Cooling of the CVT
To ensure that the produced heat in the CVT can be transported to the environment, some
modifications can be done. The air inlet in the CVT casing can be milled, to enlarge the
area of the inlet. Another option is to install a fan with another shape on the secondary
pulley, to achieve a higher airflow in the CVT casing. This modification should improve
the reliability and wear of the belt.
Connection CVT with engine
To get a compact design and low center of gravity the original engine crankcase can be
redesigned. The gearbox is not necessary any more, so the layout of the crankcase can be
altered. This can save some weight of the crankcase and the center of gravity of the
engine can be lowered. A complete new design of the crankcase can achieve these goals.
For more information, see [Modification of a Formula Student race car engine, for
addition of a Continuously Variable Transmission].
29
Chapter 13
Conclusion
In this bachelor final project the implementation of a Suzuki Electronically-controlled
Continuously Variable Transmission is researched. First, the driveline of a Suzuki
Burgman 650 is studied and an engine selection is made. An engine of a Suzuki GSX-R
600 is chosen.
The functioning of the SECVT is researched with a reversed engineering study. The
primary pulley of the CVT is electro-mechanically actuated with a DC-motor, gear train
and a spindle. The secondary pulley will follow the displacement of the primary pulley
because of a spring that is incorporated in the design of the secondary pulley.
The belt of the CVT is a dry hybrid belt (Bando AVANCE). This belt has the good
qualities of a metal and rubber belt combined in one design. Besides that the efficiency is
higher than 90% when the primary torque is more than 20 Nm.
The combination of a Suzuki GSX-R 600 engine with a SECVT might give some
problems. The forces on the belt are a lot higher in comparison to the Burgman 650
design. The manufacturer normally applies safety factors to their design. So the
overloading of the CVT will probably not cause any problems. It might appear that there
is more wear of the belt. In the design of a race car this will not be a problem, because of
the shorter lifecycle of the parts.
The shift speed of the CVT is tested on a test bench at Drivetrain Innovations. The results
of the measurements were satisfying. The CVT can shift faster than the race car can
accelerate or decelerate.
A judge of Formula Student England said: The concept of the implementation of a
SECVT in a race car is ambitious. It is a lot of work but with a good team it is possible!
University Racing Eindhoven participated in the Formula student competition in
England (Silverstone) with there concept of the URE05. With this concept they won the
first price for best design in class 3 and the first price for overall winner of class 3.
Figure 13.1: The class 3 team members of URE at the Silverstone circuit in England
30
Chapter 14
Bibliography
Suzuki AN650 service manual

Study on the efficiency of an A-CVT : developments for the constant speed power
take off , F.J.J. Gommans, 2003, Technische Universiteit Eindhoven
The constant speed power take-off (CS-PTO), D.J. de Cloe, 2003, Technische
Universiteit Eindhoven
Modification of a Formula Student race car engine, for addition of a Continuously
Variable Transmission, L. Marquenie, 2008, Technische Universiteit Eindhoven
Websites
www.neita.nl
www.bikez.com
www.ntn-europe.com
www.alpha-sports.com/suzuki_parts.htm
www.coatingservice.nl
www.contitech.de
www.contitech.de/pages/presse/presse-uebersicht/pressefachthemen/2007/071016_hybridring/presse_en.html
31
Appendix A
Section of part of Burgman 650 drive line
Figure A.1: Section of a Suzuki Burgman 650 driveline [Suzuki AN650 service manual]
32
Appendix B
M-file for calculation of power of modified

Suzuki GSX-R 600 engine
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%% Power of modified Suzuki GSXR 600 engine %%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%Auteur: Berry Kuijpers
clc
clear all
close all
%%%%%%%%%% determination of constants %%%%%%%%%%%
efficiency_drivetrain = 0.92; % assumption drivetrain of Suzuki GSX-R 600 is 92% efficient [-]
%%%%%%%%%% Measured values of modified Suzuki GSX-R 600 engine %%%%%%%%%%

P_measured
=[12;14.1;16.4;18.5;21.8;27.5;31.3;35;39;43.7;49.1;49;50.5;51.8;51.8;52.5;53;52.5;51.3;50;48;41]
; % Power of engine [kW]
rpm_crank =
[3000;3500;4000;4500;5000;5500;6000;6500;7000;7500;8000;8500;9000;9500;10000;10500;110
00;11500;12000;12500;13000;13500]; % Revolutions per minute of the engine [-]
w_crank = rpm_crank * (2*pi/60); % Angular velocity of the crank [rad/s]
%%%%%%%%%% Calculation of power of the engine ar crank %%%%%%%%%%%%%

P_crank = (1/efficiency_drivetrain)* P_measured; % Power of engine at crank [kW]
T_crank = (P_crank * 10^3) ./ w_crank; % Torque of engine at crank [Nm]
%%%%%%%%%% Figure %%%%%%%%%%%%

figure
plotyy(rpm_crank,P_crank,rpm_crank,T_crank)
axis([3000 13000 0 80]);
xlabel('RPM');
ylabel('Power [kW]');
title('Power of modified Suzuki GSX-R 600 engine at crank');
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33
Appendix C.1
Exploded view SECVT
Figure C.1: Exploded view of SECVT [Suzuki AN650 service manual]
34
Appendix C.2
Exploded view primary pulley assembly SECVT
Figure C.2: Exploded view of primary pulley assembly
Primary pulley
4
5
7
8
27
28
29
30
33
Primary pulley shaft bolt

Primary pulley shaft adaptor
Bearing
O-ring
Primary fixed pulley
Key
Plain bearing bush
Primary slide pulley gear
34
35
36
37
38
39
40
41
42
Bearing
Snap ring
Screw shaft
Bearing
Snap ring
Nut
Snap ring
Primary pulley cover
Bolt
35
Appendix C.3
Exploded view secondary pulley assembly

SECVT
Figure C.3: Exploded view of the secondary pulley assembly
Secondary pulley
1
2
13
15
17
18
19
43
Secundary pulley shaft nut

Secundary pulley shaft adaptor
Bearing
Washer
Secondary pulley fan
Bearing
Snap ring
Secondary fixed pulley
44
46
47
48
49
50
51
52

Torque cam
Spring holder cap
Snap ring
Spring baseplate
Washer
Spring (compressed)
Bolt
36
Nummer
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Naam onderdeel
Secundary pulley shaft nut
Secundary pulley shaft adaptor
O-ring
Primary pulley shaft bolt
Primary pulley shaft adaptor
CVT casing
Bearing
O-ring
CVT belt
Primary pulley assembly
Shim
O-ring
Bearing
Oil seal
Washer
Secondary pulley assembly
Bearing
Snap ring
Primary slide pulley idle gear
Secondary pulley revolution sensor
CVT motor
CVT cover
CVT filter
Primary pulley stopper bolt
Pulley position sensor
model UGS NX5.0

ja
ja
nee
ja
ja
nee
ja
ja
ja
ja
ja
ja
ja
nee
ja
ja
ja
ja
ja
ja
nee
ja
nee
nee
nee
nee
Documentation of the SECVT
Appendix C.4
Gewicht (gram)
20
135
<5
30
110
1485
50
<5
210
5
<5
75
5
<5
105
80
<5
180
50
695
2025
45
10
40
Materiaal
Staal
Staal
Rubber
Staal
Staal
Aluminium
Verschillend
Rubber
Verschillend
Staal
Rubber
Verschillend
Verschillend
Staal
Aluminium
Verschillend
Staal
Verschillend
Verschillend
Verschillend
Aluminium
Verschillend
Staal
Verschillend
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Aantal
1
1
1
1
1
1
1
1
1
37
M10x20
non-linear axial displacement sensor
inductive pulse sensor
Kogellager sealed NTN 6205LHA (25x52x15)

Circlip as 25mm
27x32x2
76x82x0,7
87x3,5
Kogellager open Koyo 60/32C3 (32x58x13)

25x1,2
72x2,5
M12x30
Specificatie
Steek 34mm
7
8
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Primary pulley assembly

Bearing
O-ring
Primary fixed pulley
Key
Plain bearing bush
Plain bearing
Plain bearing
Bearing
Snap ring
Screw shaft
Bearing
Snap ring
Nut
Snap ring
Primary pulley cover
Bolt
ja
ja
ja
ja
ja
ja
nee
nee
ja
ja
ja
ja
ja
ja
ja
ja
ja
ja
Verschillend
Rubber
Staal
Staal
Staal
Staal
Verschillend
Verschillend
Aluminium
Verschillend
Verenstaal
Staal
Verschillend
Verenstaal
Staal
Verenstaal
Staal
Staal
10
5
55
<5
505
425
170
10
1015
1260
50
<5
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3
M4x10
Circlip as 75mm
Steek 24mm
Circlip as 50mm
38
Glijlager (27x30x18)
Z6 =64
Circlip gat 80mm
Trapeziumdraad Tr64 x 14P4,67 LH
Kogellager seales NTN 6009LHA (18x75x16)

25x1,2
56
57
54
55
21
22
58
59
26
33
53
13
15
17
18
19
43
44
45
46
47
48
49
50
51
52
Actuator
Secondary pulley revolution sensor
CVT motor
CVT motor shaft
CVT motor gear
Pulley position sensor
Primary slide pulley idle gear 1
shaft
shaft
Bearing
Secondary pulley assembly

Bearing
Washer
Bearing
Snap ring
Secondary fixed pulley
Plain bearing
Torque cam
Spring holder cap
Snap ring
Spring baseplate
Washer
Spring (compressed)
Bolt
ja
ja
ja
ja
nee
ja
ja
ja
nee
ja
ja
ja
ja
ja
ja
ja
ja
ja
nee
ja
ja
ja
ja
ja
ja
ja
Staal
Verschillend
Staal
Kunststof
Verschillend
Verschillend
Staal
Kunststof
Verschillend
Aluminium
Kunststof
Verschillend
Staal
Aluminium
Verschillend
Verenstaal
Staal
Staal
Verschillend
Staal
Staal
Verenstaal
Staal
Staal
Verenstaal
Staal
50
10
20
50
40
425
20
695
50
460
40
<5
20
10
385
<5
1045
75
<5
105
80
<5
1485
1
4
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
4
39
Z4 = 35 , Z5 = 11
Z1 = 13
non-linear axial displacement sensor
Z6 = 64
Z2 = 58 , Z3 = 10
inductive pulse sensor
72x90,7x0,8
L_0 = 129 mm , k = 34,4 N/mm
M5x8
torque cam angle 32,5
18 notches
Circlip as 26mm

Circlip as 25mm
Kogellager open Koyo 60/32C3 (32x58x13)

27x32x2
Appendix C.5
Specifications of the SECVT and Bando

AVANCE belt
Specifications published by Aichi Machinery and Bando Chemical Industries
Maximum torque
Material
Width
Length
Pulley distance
Avance belt (Bando Chemical Industries)

135
Rubber/aluminum blocks
25
612
148.5
CVT Pulleys (Aichi Machine Industry Co.)

Belt ratio (low)
1.980
Belt ratio (high)
0.460
Ratio change CVT
4.304
Pitch diameter primary pulley
65.8 133.5
Pitch diameter secondary pulley
130.3 61.4
Pulley outer diameter primary pulley
147
Pulley outer diameter secondary pulley
143
Efficiency (low) (input torque > 20 Nm)
CVT > 95%
[Nm]
[-]
[mm]
[mm]
[mm]
[-]
[-]
[-]
[mm]
[mm]
[mm]
[mm]
[-]
Table C.1: Specification of the SECVT and Bando AVANCE belt
40
Appendix D
Efficiency of the SECVT at different ratios
Figure D.1: The efficiency in ratio low
Figure D.2: Sample of the efficiency in ratio low
Figure D.3: The efficiency in ratio 1:1
Figure D.4: Sample of the efficiency in ratio 1:1
Figure D.5: The efficiency in ratio high
Figure D.6: Sample of the efficiency in ratio high
Remark: The plotted lines in figures D.2, D.4 and D.6 serve as a guide to the eye
41
Appendix E.1
Torque on belt with different reductions

When the SECVT is connected to the Suzuki GSX-R 600 engine, a primary reduction is
necessary to ensure a good lifetime of the belt. In the following figures different setups
are shown. The setup with the primary reduction and fifth gear is the setup which is close
to the ideal reduction. The M-file with the calculations of the forces on the belt is shown
in appendix E.2.
Torque on CVT belt with primary reduction and 5th gear

200
180
160
140
120
100
80
60
40
Torque on belt
20
2000
2500
3000
3500
4000
4500
5000
5500
rpm primary pulley
42
Torque on CVT belt with primary reduction and 6th gear

200
180
160
140
120
100
80
60
40
Torque on belt
20
2000
2500
3000
3500
4000
4500
5000
5500
6000
rpm primary pulley
Torque on CVT belt with primary reduction

200
180
160
140
120
100
80
60
40
Torque on belt
20
2000
2500
3000
3500
4000
4500
5000
5500
6000
rpm primary pulley
43
Appendix E.2
M-file of the calculations of the forces on the

CVT belt
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%% Determination of forces on the CVT belt %%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
clc
clear all
close all
i_primair_gsx600 = 79/41; % Primary reduction Suzuki GSX-R 600 (out/in) [-]
gear_transmission_4 = 30/20; % Gear ratio of 4th gear (out/in) [-]
efficiency_drivetrain = 0.92; % assumption drivetrain of Suzuki GSX-R 600 is 92% efficient [-]
max_rpm_engine = 13000; % maximum revolutions per minute of the engine [-]
max_v_belt = 40; % maximum velocity of the CVT belt [m/s]
r_primary_pulley_min = 0.0329; % minimum pitch radius of primary pulley CVT [m]
r_primary_pulley_max = 0.06677; % maximum pitch radius of primary pulley CVT [m]
r_primary_pulley = [r_primary_pulley_min:0.0001:r_primary_pulley_max]'; % pitch radius on
primary pulley CVT [m]
%%%%%%%%%% Measured values of modified Suzuki GSX-R 600 engine %%%%%%%%%%

P_measured
=[12;14.1;16.4;18.5;21.8;27.5;31.3;35;39;43.7;49.1;49;50.5;51.8;51.8;52.5;53;52.5;51.3;50;48;41]
* 10^3; % Power of engine [W]
rpm_crank =
[3000;3500;4000;4500;5000;5500;6000;6500;7000;7500;8000;8500;9000;9500;10000;10500;110
00;11500;12000;12500;13000;13500]; % Revolutions per minute of the engine [-]
w_crank = rpm_crank * (2*pi/60); % Angular velocity of the crank [rad/s]
%%%%%%%%%% Calculation of power of the engine ar crank %%%%%%%%%%%%%

P_crank = (1/efficiency_drivetrain)* P_measured; % Power of engine at crank [W]
T_crank = P_crank ./ w_crank; % Torque of engine at crank [Nm]
44
%%%%%%%%%% Calculation of ideal primary reduction %%%%%%%%%%%

max_w_engine = max_rpm_engine * ((2*pi)/60); % Maximum angular velocity of the engine
[rad/s]
i_primair_perfect = max_w_engine / (max_v_belt / r_primary_pulley_max); % Ideal primary
reduction ratio [-]
%%%%%%%%%% Calculation of variables %%%%%%%%%%%

% Primary reduction [-]
i_primair = [i_primair_gsx600
i_primair_gsx600 * gear_transmission_5
i_primair_gsx600 * gear_transmission_6
i_primair_perfect];
% Angular velocity primary pulley CVT [rad/s]
w_cvt_primary_1 = w_crank / i_primair(1);
% Revolutions per minute of primary pulley CVT [-]
rpm_cvt_primary_1 = w_cvt_primary_1 * (60/(2*pi));
% Torque on primary pulley CVT [Nm]
T_cvt_primary_1 = T_crank * i_primair(1);
% Velocity of CVT belt on primary pulley CVT [m/s]
v_belt_1 = w_cvt_primary_1(21) * r_primary_pulley; %snelheid van de band (m/s)
v_belt_2 = w_cvt_primary_2(21) * r_primary_pulley;
Bando_Avance = 135 * ones(1,length(T_cvt_primary_1));
Contitech_Hybrid_Ring = 180 * ones(1,length(T_cvt_primary_1));
v_max = max_v_belt * ones(1,length(r_primary_pulley));
45
%%%%%%%%% Figures %%%%%%%%%%%

% Velocity of CVT belt on primary pulley CVT
figure
plot(r_primary_pulley,v_max,'r','LineWidth',2);
hold on
plot(r_primary_pulley,v_belt_1,'b','LineWidth',2);
plot(r_primary_pulley,v_belt_2,'g','LineWidth',2);
plot(r_primary_pulley,v_belt_3,'k','LineWidth',2);
plot(r_primary_pulley,v_belt_4,'m','LineWidth',2);
axis([r_primary_pulley_min r_primary_pulley_max 10 60]);
xlabel('radius primary pulley [m]');
ylabel('speed belt [m/s]');
title('Speed of the CVT belt @ 13000 RPM');
legend('max allowable speed','with primary reduction','with primary reduction and 5th gear','with
primary reduction and 6th gear','with ideal reduction');
hold off
% Torque on CVT belt with ideal primary reduction
figure
plot(rpm_cvt_primary_4,T_cvt_primary_4,'b','LineWidth',2);
hold on
plot(rpm_cvt_primary_4,Bando_Avance,'r','LineWidth',2);
plot(rpm_cvt_primary_4,Contitech_Hybrid_Ring,'k','LineWidth',2);
axis([1600 6300 0 200]);
xlabel('rpm primary pulley');
ylabel('Torque on CVT belt [Nm]');
title('Torque on CVT belt with ideal reduction');
legend('Torque on belt','Max torque Bando Avance','Max torque Contitech Hybrid Ring');
hold off
% Torque on CVT belt with original primary reduction Suzuki GSX-R 600 engine
figure
hold on
axis([1600 6300 0 200]);
title('Torque on CVT belt with primary reduction');
hold off
% Torque on CVT belt with original primary reduction and 5th gear of Suzuki GSX-R 600 engine
figure
hold on
axis([1600 5700 0 200]);
title('Torque on CVT belt with primary reduction and 5th gear ');
46
hold off
% Torque on CVT belt with original primary reduction and 6th gear of Suzuki GSX-R 600 engine
figure
hold on
axis([1600 6300 0 200]);
title('Torque on CVT belt with primary reduction and 6th gear');
hold off
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
47
Appendix F.1
Thrust force calculations

The thrust force as is applied to the belt at the secondary pulley is displayed in figure 8.1.
These values can be calculated using the following formulas. First the force applied by
the spring is calculated (equation F.1). The second step is to calculate the force applied
by the torque-cam mechanism (equation F.2). This force can not be lower than zero thus
the outcome of equation F.2 has to be altered in case of Fcam < 0.
When combining Fspring and Fcam, the total thrust force Fthrust can be calculated using
equation F.3.
Fspring = k spring ( x0 + xspring )
T primair belt
Fcam =
Fthrust =
Variable
kspring
x0
xspring
Tprimair
belt
rcvt
s
Rspring
Rcam
2 rcvt
Equation F.1
Fspring s Rspring
Rcam
Fcam
+ Fspring
tan( + )
Equation F.2
Equation F.3
Value [min/max]
34.4 N/mm (measured at Technoflex)
68.52 mm
0 / 15.7 mm = 2 * Rsec * tan()
0 / 149 Nm
0.95 [-]
0.55 / 2.05 [-]
0.12 [-]
40.75 mm
30.85 mm
32.5
(tan(c)) -1
0.18 [-]
13
Table F.1 Values for calculation of thrust force [Study on the efficiency of an A-CVT :
developments for the constant speed power take off , F.J.J. Gommans, 2003, Technische
Universiteit Eindhoven]
* Variables are explained in M-file in appendix D??
48
Appendix F.2
M-file for calculation of thrust force

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%% Calculation thrustforce SE-CVT with torquecam %%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
clc
clear all
close all
k_spring = 34400; % Spring rate of compression spring (measured at Technoflex) [N/m]
x_0 = 0.06852; % Initial setting length of spring [m]
efficiency_belt = 0.95; % Efficiency of the CVT belt [-]
r_cvt = [0.55:0.01:2.05]'; % Ratio coverage of SECVT [-]
delta_x_spring = [0:(0.0157/length(r_cvt)):0.0157]'; % Change of spring length [m]
mu_s = 0.12; % Coefficient of friction on spring seat [-]
R_spring = 0.04075; % Radius of spring (m)
R_cam = 0.03085; % Radius of torquecam (m)
labda = 32.5; % Cam angle [degrees]
mu_c = 0.18; % Coefficient of friction on cam [-]
alpha = atand(mu_c); % Friction angle on cam [degrees]
max_T_engine = 65.67; % Maximum torque of modified Suzuki GSX-R 600 engine [Nm]
i_primair_perfect = 2.2724; % Ideal primary reduction ratio of modified Suzuki GSX-R 600 engine
with SECVT [-]
max_T_engine_primair = max_T_engine * i_primair_perfect; % Torque of modified Suzuki GSX-R
600 engine at input shaft SECVT [Nm]
max_T_Burgman = 74; % Maximum torque of Suzuki Burgman 650 at imput shaft SECVT [Nm]
T_primair =[0 20 40 max_T_Burgman max_T_engine_primair]; % Torque at input shaft of SECVT
[Nm]
%%%%%%%%%%% Calculation of thrust force %%%%%%%%%%

figure % Open figure
for k = 1 : length(T_primair)
for n = 1 : length(r_cvt)
F_spring(n) = k_spring * (x_0 + delta_x_spring(n)); % Spring force [N]
F_cam(n) = (((T_primair(k) * efficiency_belt)/(2 * r_cvt(n))) - (F_spring(n) * mu_s * R_spring))
/ R_cam; % Tangential force on cam [N]
49
% In the case the tangential force on cam is below zero the output is altered
if F_cam(n) < 0
F_cam(n) = 0;
end
F_thrust(n) = (F_cam(n) / (tand(labda + alpha))) + F_spring(n); % Thrust force generated by
cam and spring [N]
end
plot(r_cvt,F_thrust,'b','LineWidth',2); % Plot outcome in figure
hold on
end
%%%%%%%%%%% Figure properties %%%%%%%%%%%%%%%%%%%

axis([0.55 2.05 0 7000]); % Define axis properties
xlabel('CVT ratio [-]');
ylabel('Thrustforce [N]');
title('Thrustforce secondary pulley SE-CVT with torque cam');
legend('0 Nm','20 Nm','40 Nm','74 Nm','149 Nm');
hold off
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
50
Appendix G.1
Manufacturing drawing of adapter for test bench
Figure M.1: Manufacturing drawing of adapter for test bench, manufactured at VDL
51
Appendix G.2
Wiring scheme of the test set up

10 A
12 V
Relais
10 k
Power supply
Vout1
Resistor
Transistor
D-space
DC motor
10 A
12 V
Relais
Power supply
SECVT
Power
PPS
Ground
SECVT
5V
Vin1
Power supply
COAX
5 V signal
D-space
Signal
Ground
12 V
REV Sec
Amplifier
Power
SECVT
8 Gnd
inc1
Power supply
Signal
6 ipx (11)
D-space
Pull up resistor
2 k
1 VCC
Figure N.1: Wiring scheme of the test set up at DTI
52
Appendix G.3
M-file for calibration of the pulley position

sensor
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%% Pulley Position Sensor Calibration file %%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%Auteur: Harrie Smetsers
%This file has been used to calibrate the pulley position sensor. A micro
%meter is taken and the sensor is compressed and every millimiter the
%output voltage is recorded. This measurement has been done 3 times and an
%average is taken.
clc
clear all
close all
%Measured data
pps_sensor = [0.1420 0.2340 0.3563 0.4847 0.6277 0.7650 0.9017 1.0500 1.1933 1.3433 1.5367
1.7467 1.9700 2.1900 2.4333 2.7233 3.0400 3.3967 3.8033 4.3000 4.8200;
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16
17 18 19 20];
%Polyfit
p1 = pps_sensor;
n= 4;
p2=polyfit(p1(1,:),p1(2,:),n);
x = 0:0.01:5; f = p2(1)*x.^4 + p2(2)*x.^3 + p2(3)*x.^2 + p2(4)*x + p2(5)
%Figure
figure(1);
orient tall
plot(p1(1,:),p1(2,:),'k--');
hold on;
plot(x,f,'r');
ylabel('position [mm]');
xlabel('voltage [V]');
x1 = line(xlim,[3 3]);
x2 = line(xlim,[18 18]);
set(x1,'LineWidth',1,'LineStyle','-.','Color','k')
set(x2,'LineWidth',1,'LineStyle','-.','Color','k')
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
53
Appendix G.4
M-file for data processing of the measurement

data while downshifting at 1000 rpm
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%% File to plot figures with the different measuring data %%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%% Auteurs: Harrie Smetsers and Berry Kuijpers
clear all
close all
clc
%%%% determination of constants %%%%
r_primary_max = 66.75;
r_primary_min = 32.90;
r_secondary_max = 65.15;
r_secondary_min = 30.70;
beta = 13;
%Maximum pitch radius primary pulley CVT [mm]

%Minimum pitch radius primary pulley CVT [mm]
%Maximum pitch radius secondary pulley CVT [mm]
%Minimum pitch radius secondary pulley CVT [mm]
%Angle of pulley sheave with vertical [degrees]
%%%% polyline for converting pps sensor voltage to position %%%%

pps = [0.1420 0.2340 0.3563 0.4847 0.6277 0.7650 0.9017 1.0500 1.1933 1.3433 1.5367 1.7467
1.9700 2.1900 2.4333 2.7233 3.0400 3.3967 3.8033 4.3000 4.8200;
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17
18 19 20];
n= 4;
pps_fit=polyfit(pps(1,:),pps(2,:),n);
%%%% Load data measurements downshift with primary pulley at 1000 rpm %%%%
%%%% Measurement 1 %%%%
%Load data
load 1000rpmup_down_shift\test1000rpm1down.mat
t1 = trace_x;
%time
V1 = trace_y(1,:);
%trigger signal of the shift motor
p1 = trace_y(2,:);
%pulley position sensor [mm]
P1 = trace_y(3,:);
%pulley position sensor [V]
% calculation rpm primary and secondary pulley
np1 = ones(1,length(t1)).*1000;
%rpm primary pulley [-]
rprim1 = r_primary_max - (p1 ./ (2 * tand(beta))); %radius primary pulley [mm]
for n = 1 : length(rprim1)
if rprim1(n) < r_primary_min
%correction for outer values
rprim1(n) = r_primary_min;
54
end
end
rsec1 = r_secondary_min + (p1 ./ (2 * tand(beta))); %radius secondary pulley [mm]
for n = 1 : length(rsec1)
if rsec1(n) > r_secondary_max
rsec1(n) = r_secondary_max;
end
end
ns1 = np1 .* (rsec1./rprim1);
%rpm secondary pulley [-]
% Trigger time signal
z1 = find(trace_y(1,:)>2.5);
%index for trigger signal on 0.5s
t1star = t1-t1(z1(1,1))+0.51;
%new time scale with trigger on 0.5s
t1 = t1star;
%rewrite time

%Load data
t2 = trace_x;
%time
V2 = trace_y(1,:);
p2 = trace_y(2,:);
P2 = trace_y(3,:);
end
end
end
end
z2 = find(trace_y(1,:)>2.5);
t2star = t2-t2(z2(1,1))+0.51;
t2 = t2star;
%rewrite time

%Load data
t3 = trace_x;
%time
V3 = trace_y(1,:);
p3 = trace_y(2,:);
P3 = trace_y(3,:);
55

end
end
end
end
z3 = find(trace_y(1,:)>2.5);
t3star = t3-t3(z3(1,1))+0.51;
t3 = t3star;
%rewrite time

%Load data
t4 = trace_x;
%time
V4 = trace_y(1,:);
p4 = trace_y(2,:);
P4 = trace_y(3,:);
end
end
end
end
z4 = find(trace_y(1,:)>2.5);
t4star = t4-t4(z4(1,1))+0.51;
t4 = t4star;
%rewrite time

%Load data
56
t5 = trace_x;
%time
V5 = trace_y(1,:);
p5 = trace_y(2,:);
P5 = trace_y(3,:);
end
end
end
end
z5 = find(trace_y(1,:)>2.5);
t5star = t5-t5(z5(1,1))+0.51;
t5 = t5star;
%rewrite time
%%%% figures %%%%

% trigger signal
figure(1);
orient tall
plot(t1,V1,'k--',t2,V2,'b--',t3,V3,'r--',t4,V4,'g--',t5,V5,'m--');
axis([0 4 0 6])
ylabel('voltage [V]');
xlabel('time [s]');
legend('m1','m2','m3','m4','m5')
set(legend,'Location','NorthWest');
% rpm signal
figure(2);
orient tall
plot(t1,np1,'k:',t1,ns1,'k-.');
hold on
plot(t2,np2,'b:',t2,ns2,'b-.');
plot(t3,np3,'r:',t3,ns3,'r-.');
plot(t4,np4,'g:',t4,ns4,'g-.');
plot(t5,np5,'m:',t5,ns5,'m-.');
axis([0 4 0 2500])
ylabel('rpm [-]');
xlabel('time [s]');
legend('primary pulley','secondary pulley');
title('Revolutions per minute of the primary and secondary pulley while downshifting at 1000
rpm');
57
% ppsensor
figure(3);
orient tall
plot(t1,p1,'k-.',t1,P1,'b-.'); hold on
plot(t2,p2,'k--',t2,P2,'b--')
plot(t3,p3,'k--',t3,P3,'b--')
plot(t4,p4,'k--',t4,P4,'b--')
plot(t5,p5,'k--',t5,P5,'b--')
axis([0 4 0 22])
ylabel('compression [mm] / voltage [V]');
xlabel('time [s]');
legend('compression','voltage');
title('Compression and voltage of the pulley position sensor while downshifting at 1000 rpm');
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
58
Appendix G.5
Test results of measurement shift speed at 2000

and 3000 rpm
22
compression
voltage
20
18
16
14
12
10
0.5
1.5
2.5
3.5
time [s]
5000
primary pulley}
secondary pulley
4500
4000
3500
rpm [-]
3000
2500
2000
1500
1000
500
0.5
1.5
2.5
3.5
time [s]
59
22
compression
voltage
20
18
16
14
12
10
0.5
1.5
2.5
3.5
time [s]
5000
primary pulley
secondary pulley
4500
4000
3500
rpm [-]
3000
2500
2000
1500
1000
500
0.5
1.5
2.5
3.5
time [s]
60
22
compression
voltage
20
18
16
14
12
10
10
time [s]
8000
primary pulley
secondary pulley
7000
6000
rpm [-]
5000
4000
3000
2000
1000
10
time [s]
61
22
compression
voltage
20
18
16
14
12
10
10
time [s]
8000
primary pulley
secondary pulley
7000
6000
rpm [-]
5000
4000
3000
2000
1000
10
time [s]
62

SECVT Suzuki Motos

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

SECVT Suzuki Motos

Hochgeladen von

Copyright:

Verfügbare Formate

Implementation of a Suzuki Electronicallycontrolled Continuously Variable

Transmission in a Formula Student racecar

Bachelor Final Project

2 Suzuki Burgman 650

6 Efficiency of the SECVT

9 Cooling of the SECVT

10 Control of the CVT

12 Recommendations to achieve higher performance

A Section of part of Burgman 650 drive line

B M-file for calculation of power of modified Suzuki GSX-R 600 engine

C.1 Exploded view SECVT

C.2 Exploded view primary pulley assembly SECVT

C.3 Exploded view secondary pulley assembly SECVT

C.4 Documentation of the SECVT

C.5 Specifications of the SECVT and Bando AVANCE belt

D Efficiency of the SECVT at different ratios

E.1 Torque on belt with different reductions

F.1 Thrust force calculations

F.2 M-file for calculation of thrust force

G.1 Manufacturing drawing of adapter for test bench

G.2 Wiring scheme of the test set up

G.3 M-file for calibration of the pulley position sensor

G.4 M-file for data processing of the measurement data

Figure 1.1: Suzuki Electronically-controlled Continuously Variable Transmission

Suzuki Burgman 650

2.1 Driveline Suzuki Burgman 650

3.1 Suzuki Burgman 650 engine

Four-stroke, Liquid-cooled, DOHC

Table 3.1: Specifications of a Suzuki Burgman 650 engine [ www.bikez.com]

3.2 Suzuki GSX-R 600 engine

Modified restricted version

Table 3.2: Specifications of a Suzuki GSX-R 600 engine [ www.bikez.com]

Figure 3.1: Figure of modified Suzuki GSX-R 600

The advantages of this engine are:

3.3 Engine choice

Suzuki Electronically-controlled Continuously

Figure 4.1: Top view of SECVT

Figure 4.2: Primary pulley assembly (open)

Figure 4.3: Secondary pulley assembly (closed)

4.2 Ratio changing mechanism

Figure 4.4: Actuation mechanism SECVT

Figure 4.5: Torque-cam mechanism secondary pulley SECVT

4.3 Degrees of freedom (DOF)

Primary slide pulley

Primary slide pulley gear (when actuated)

Secondary slide pulley

3D CAD model of the SECVT

Figure 5.1: Side views of 3D CAD model of SECVT

Efficiency of the SECVT

6.1 Measurement of the efficiency

Pout = sec ondair Tsec ondair

Figure 6.1: Efficiency of SECVT at ratio low

6.2 Race car application

Figure 7.1: Section of the Bando AVANCE belt

7.1 Bando AVANCE

7.2 Limitations of the belt

speed belt [m/s]

radius primary pulley [m]

Figure 7.2: Speed of the CVT belt at different primary reductions

Torque on CVT belt with ideal reduction

Torque on CVT belt [Nm]

rpm primary pulley

7.3 Contitech Hybrid Ring

Figure 7.4: Contitech Hybrid Ring